18F-FDG PET/CT (PET/CT) influences management in patients with known or suspected pancreatic cancer

  • Dr Thomas Barber, The Alfred Hospital, Australia
  • A/Prof Victor Kalff, Department of Nuclear Medicine ,The Alfred Hospital, Melbourne, Victoria, Australia
  • Dr Martin Cherk, Department of Nuclear Medicine ,The Alfred Hospital, Melbourne, Victoria, Australia
  • Dr Kenneth Yap, Department of Nuclear Medicine ,The Alfred Hospital, Melbourne, Victoria, Australia
  • Mr Peter Evans, Department of Surgery, The Alfred Hospital, Melbourne, Victoria, Australia
  • A/Prof Michael Kelly, Department of Nuclear Medicine ,The Alfred Hospital, Melbourne, Victoria, Australia

    • Objective: To assess the impact on clinical management of PET/CT in patients with known or suspected pancreatic cancer.

    Methods: Between April 2006 and September 2008, 25 PET/CT scans were performed using a dedicated PET/CT (22 scans) or a coincidence hybrid PET/CT camera (3 scans) in 23 patients with known or suspected pancreatic cancer. 17 scans were performed for initial evaluation and 8 for restaging of disease. The pre-PET/CT management plan and/or intent were prospectively recorded in all cases. The post-PET/CT management plan was determined from the medical record and/or discussions with treating clinicians. The impact of PET/CT on management was classified as High, Medium, Low or None, defined using ANZAPNM PET data collection project criteria. Follow-up was used to reconcile any discordance between PET/CT and conventional imaging.

    Results: Overall, PET/CT management impact was classified as high (n=7), medium (n=4), low (n=10) or none (n=4). Impact was either high or medium in 11/25 patients (44%) (95% confidence interval; 24 – 64%). Impact was high in 4/17 patients imaged for initial evaluation, predominantly by clarifying equivocal lesions on conventional imaging. In restaged patients, PET/CT impact was high in 3/8, and it correctly modified disease extent in 5/8. In the 16 discordant studies, PET/CT assessment was correct in 10, conventional imaging in 4 and there was insufficient information in 2.

    Conclusion: PET/CT has high or medium management impact in 44% of patients imaged for known or suspected pancreatic cancer, more commonly during restaging. Discordant PET/CT results were usually correct.